Open Access Discussions Global impact of smoke aerosols ... · M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5229 above the freezing level is

Atmos. Chem. Phys., 13, 5227–5241, 2013www.atmos-chem-phys.net/13/5227/2013/doi:10.5194/acp-13-5227-2013© Author(s) 2013. CC Attribution 3.0 License.

EGU Journal Logos (RGB)

Advances in Geosciences

Open A

ccess

Natural Hazards and Earth System

Sciences

Open A

ccess

Annales Geophysicae

Open A

ccessNonlinear Processes

in Geophysics

Open A

ccess

Atmospheric Chemistry

and PhysicsO

pen Access

Atmospheric Chemistry

and Physics

Open A

ccess

Discussions

Atmospheric Measurement

Techniques

Open A

ccess

Atmospheric Measurement

Techniques

Open A

ccess

Discussions

Biogeosciences

Open A

ccess

Open A

ccess

BiogeosciencesDiscussions

Climate of the Past

Open A

ccess

Open A

ccess

Climate of the Past

Discussions

Earth System Dynamics

Open A

ccess

Open A

ccess

Earth System Dynamics

Discussions

GeoscientificInstrumentation

Methods andData Systems

Open A

ccess

GeoscientificInstrumentation

Methods andData Systems

Open A

ccess

Discussions

GeoscientificModel Development

Open A

ccess

Open A

ccess

GeoscientificModel Development

Discussions

Hydrology and Earth System

Sciences

Open A

ccess

Hydrology and Earth System

Sciences

Open A

ccess

Discussions

Ocean Science

Open A

ccess

Open A

ccess

Ocean ScienceDiscussions

Solid Earth

Open A

ccess

Open A

ccess

Solid EarthDiscussions

The Cryosphere

Open A

ccess

Open A

ccess

The CryosphereDiscussions

Natural Hazards and Earth System

Sciences

Open A

ccess

Discussions

Global impact of smoke aerosols from landscape fireson climate and the Hadley circulation

M. G. Tosca1,*, J. T. Randerson1, and C. S. Zender1

1Department of Earth System Science, University of California, Irvine, CA 92697, USA* now at: NASA Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA

Correspondence to:M. G. Tosca ([email protected])

Received: 1 October 2012 – Published in Atmos. Chem. Phys. Discuss.: 25 October 2012Revised: 12 April 2013 – Accepted: 24 April 2013 – Published: 24 May 2013

Abstract. Each year landscape fires across the globe emitblack and organic carbon smoke particles that can last inthe atmosphere for days to weeks. We characterized theclimate response to these aerosols using an Earth systemmodel. We used remote sensing observations of aerosol opti-cal depth (AOD) and simulations from the Community At-mosphere Model, version 5 (CAM5) to optimize satellite-derived smoke emissions for high biomass burning regions.Subsequent global simulations using the adjusted fire emis-sions produced AODs that were in closer agreement with sur-face and space-based measurements. We then used CAM5,which included radiative aerosol effects, to evaluate the cli-mate response to the fire-aerosol forcing. We conducted two52 yr simulations, one with four sets of monthly cycling1997–2009 fire emissions and one without. Fire emissionsincreased global mean annual AOD by 10 % (+0.02) and de-creased net all-sky surface radiation by 1 % (1.3 W m−2).Elevated AODs reduced global surface temperatures by0.13± 0.01◦C. Though global precipitation declined onlyslightly, patterns of precipitation changed, with large reduc-tions near the Equator offset by smaller increases north andsouth of the intertropical convergence zone (ITCZ). A com-bination of increased tropospheric heating and reduced sur-face temperatures increased equatorial subsidence and weak-ened the Hadley circulation. As a consequence, precipita-tion decreased over tropical forests in South America, Africaand equatorial Asia. These results are consistent with theobserved correlation between global temperatures and thestrength of the Hadley circulation and studies linking tro-pospheric heating from black carbon aerosols with tropicalexpansion.

1 Introduction

Climate is a primary driver of global and regional fire ac-tivity, and fires, in turn, influence climate on similar tempo-ral and spatial scales by means of emissions of trace gasesand aerosols and by modifying vegetation composition andstructure (Marlon et al., 2008; Power et al., 2008; Bow-man et al., 2009; Ward et al., 2012). Fire incidence was lowoutside of the tropics and subtropics during the last glacialmaximum, coinciding with cool temperatures, but increasedas global temperatures began rising around 12 000 yr ago(Power et al., 2008). During the last two millennia, fires de-creased between AD 1 and 1750, during a period of gradualglobal cooling (Marlon et al., 2008). Subsequently, between1750 and 1870, fire activity, inferred from charcoal records,rapidly increased, coinciding with a period of temperature in-creases but also when humans began exerting greater controlon ecosystem processes through land management (Marlonet al., 2008). In the American Southwest, regionally largefire years over the last several centuries often followed drywinters preceded by several years of cool and wet condi-tions that allowed fuels to accumulate (Swetnam and Betan-court, 1998). In western North America, anthropogenic cli-mate warming over the last several decades has increased thenumber of large wildland fires (Westerling et al., 2006) andalso may have influenced burn severity and levels of fuel con-sumption (Turetsky et al., 2011). On interannual timescales,satellite observations of burned area and active-fire thermalanomalies provide evidence that the El Nino–Southern Oscil-lation and other climate modes modify fire activity consider-ably in tropical forest and savanna ecosystems (Spessa et al.,2005; van der Werf et al., 2008; Field et al., 2009; Fernandeset al., 2011; Chen et al., 2011).

Published by Copernicus Publications on behalf of the European Geosciences Union.

5228 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

Feedback between climate and fires is possible becausefires also modify climate through several different pathways.Fires have contributed to the accumulation of carbon diox-ide and methane in the atmosphere in recent decades, forexample, by enabling more rapid rates of land clearing inforest ecosystems (Crutzen et al., 1979; Langenfelds et al.,2002; Page et al., 2002). Fire emissions from deforestationfires were approximately 490 Tg C yr−1 during 1997–2009(van der Werf et al., 2010). This flux, which equals a quar-ter of all global fire emissions, represents a net source ofCO2 because many forests are being permanently replacedby pastures and croplands. Fires have also contributed totropical forest degradation during this period (i.e., the loss oftrees and biomass in nearby forests not intentionally cleared),and although this flux is difficult to quantify, it likely repre-sents another important source of carbon emissions (Mortonet al., 2011). Fires also influence climate by inducing veg-etation mortality, with longer-term effects on surface albedoand energy exchange as a consequence of post-fire vegetationsuccession (Myhre et al., 2005; Lyons et al., 2008; Liu andRanderson, 2008). Emissions of ozone precursors may haveimmediate consequences for radiative forcing (Ward et al.,2012) and also longer-term effects on canopy conductanceand ecosystem carbon storage (Sitch et al., 2007).

In this study, we focus on another important climate driver:emissions of smoke aerosols. While the radiative effects ofsmoke aerosols from fires have been investigated for sev-eral decades (Kaufman et al., 1991; Penner et al., 1992;Chylek and Wong, 1995; Hansen et al., 1997; Ramanathanand Carmichael, 2008), as described below, important un-certainties remain with respect to the temporal and spatialmagnitude of surface and top-of-atmosphere forcing causedby smoke from landscape fires (Reid et al., 2009). Even lessis known about how this forcing subsequently modifies at-mosphere and surface energy fluxes, cloud lifetimes, circu-lation characteristics, and regional to global scale tempera-ture and precipitation patterns. Here we investigate the rela-tionship between forcing and climate response to fires usinga global Earth system model that includes direct and semi-direct aerosol effects. In the remainder of the introductionwe review recent work on smoke aerosol radiative forcingand relevant processes influencing large-scale climate inter-actions.

Black and organic carbon (BC and OC) are primary con-stituents of smoke aerosols from landscape fires, with BCaccounting for 5–10 % of the total particle mass and OC ac-counting for much of the remainder (Andreae and Merlet,2001; Reid et al., 2005). Mahowald et al.(2011) estimatethat approximately 60 Tg of smoke is emitted from landscapefires each year. This constitutes 30 % of the total black andorganic smoke mass emitted globally each year (Lamarqueet al., 2010). These aerosols alter the climate through thescattering and absorption of solar radiation, which simultane-ously cools the surface and warms the atmospheric column(Penner et al., 1992; Hansen et al., 1997; Ramanathan and

Carmichael, 2008), and by modifying cloud properties (Pen-ner et al., 1992; Ackerman et al., 2000). Bauer and Menon(2012) estimate that the direct radiative effect of smoke fromgrass fires, forest fires and agricultural waste burning is closeto zero globally. This forcing, however, is the residual oflarger regional and seasonal warming and cooling terms, withnegative fluxes in tropical land and ocean regions and pos-itive fluxes in polar regions.Jones et al.(2007) estimatedthe direct global radiative forcing from fire aerosols to be−0.29 W m−2, leading to a global mean temperature de-crease of 0.25◦C in the Hadley Centre climate model and aforcing efficacy of 0.86.

Accumulating evidence suggests that smoke-inducedchanges in net column shortwave radiation and interactionsbetween smoke particles and cloud droplets can modify pre-cipitation (Andreae et al., 2004; Rosenfeld, 2006; Rosenfeldet al., 2008; Andreae and Rosenfeld, 2008). Widespread con-vection suppression, the result of lowered surface tempera-tures and elevated atmospheric heating via BC absorption,increases vertical stratification, which inhibits both cloudformation and precipitation (Ackerman et al., 2000; Fein-gold et al., 2001; Tosca et al., 2010). Including smoke inclimate simulations over the Amazon caused a change inmonsoonal circulation in regions with aerosol optical depthsgreater than 0.3 (Zhang et al., 2009). In the Zhang et al.(2009) study, smoke heating increased surface pressure, de-creased upward vertical velocity and reduced the lapse rate,the combination of which increased surface divergence. Asa consequence, the onset of early autumn monsoonal rainswas delayed. Analysis of satellite observations byKorenet al.(2004) provides support for this mechanism: areas withthick smoke over the Amazon had fewer clouds. The en-trainment of microscopic smoke particles into clouds alsoacts to suppress precipitation by slowing the conversion ofcloud drops into raindrops (Gunn and Phillips, 1957; Rosen-feld et al., 2008). Using satellite observations from the Tropi-cal Rainfall Measuring Mission (TRMM) of smoke-pollutedclouds over the Amazon,Rosenfeld(1999) detected amplewater for rainfall, but a lack of precipitation due to numeroussmall water droplets. In contrast to the semi-direct aerosoleffect described byAckerman et al.(2000), where smoke-induced radiative heating limits the formation of trade cumu-lus clouds,Albrecht (1989) provided evidence that aerosolsin marine stratocumulus regions increase cloudiness and de-crease cloud droplet sizes, effectively limiting drizzle. Con-sidering all of these effects together, contemporary aerosols,including smoke from landscape fires, likely weaken the hy-drologic cycle (Ramanathan et al., 2001). Recent increasesin tropical aerosols over the last half century from anthro-pogenic activity (Field et al., 2009) may offset the expectedstrengthening of the hydrologic cycle from global warming(Held and Soden, 2006).

In some areas, ingestion of smoke aerosols into ice-phasecumulonimbus clouds may increase local precipitation. Insmoke-polluted cumulus clouds, the percentage of droplets

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5229

above the freezing level is larger, which maximizes the life-time and vertical size of the cloud and increases the inten-sity of downdrafts and precipitation rates (Rosenfeld et al.,2008). Koren et al.(2005) observed invigoration of convec-tive clouds by biomass burning aerosols over the AtlanticOcean. Taken together, these studies illustrate the uncer-tainties involved in understanding fire aerosol effects at theglobal scale. However, the inclusion of improved moist tur-bulence schemes and better representation of aerosol–cloudmicrophysical interactions in Earth system models (Brether-ton and Park, 2009) provide unique opportunities to examinefire aerosol effects on regional and global climate.

Recent work suggests that the mean strength of the Hadleycirculation is increasing (Mitas and Clement, 2005), andthough most studies attribute this strengthening and expan-sion to higher surface temperatures (Lu et al., 2007; Quanet al., 2005), there is evidence that aerosols, especially blackcarbon and sulfate, play a role in altering the mean circu-lation (Yoshimori and Broccoli, 2009). Jones et al.(2007)suggest that increased atmospheric loading of biomass burn-ing aerosols shifts the location of the inter-tropical conver-gence zone, andAllen et al. (2012b) argue that black car-bon aerosol forcing helps explain the seasonality and extentof recent Hadley cell expansion. Specifically, black carbon-induced heating of the lower troposphere at mid-latitudes sig-nificantly contributes to the observed poleward shift of thedescending branch of the Hadley circulation (Allen et al.,2012a,b). Our work isolates the impact of fire aerosols onmean global circulation patterns using an Earth system modelthat includes direct and semi-direct aerosol effects. In addi-tion, we quantify the impact of smoke aerosols on climatevariables intrinsically linked to precipitation and radiationchanges.

2 Methods

We used an Earth system model with interactive atmosphericchemistry to simulate climate with and without landscapefire aerosols. We first optimized black carbon (BC) and or-ganic carbon (OC) emissions from fire by matching simu-lated aerosol optical depths (AODs) to observations and scal-ing emissions by regionally unique factors that best matchedobserved AODs in high biomass burning regions. We thenperformed two 52 yr simulations with and without the ad-justed fire aerosol emissions and assessed the impact of theseaerosols on global temperature, precipitation and the meanHadley circulation.

2.1 Model and data description

For our simulations we used the Community Earth SystemModel (CESM), version 1, initialized with the CommunityAtmosphere Model, version 5 (CAM5), and the single-layerocean model (SOM) (Neale et al., 2010). The full chem-istry model embedded in CAM5 for this experiment wasthe Model for Ozone and Related Chemical Tracers, ver-sion 4 (MOZART-4) (Emmons et al., 2010). Like previousversions of CAM (CAM3 and CAM4), this configuration(trop mozart) includes direct and semi-direct aerosol radia-tive effects (Collins et al., 2004) and utilizes the bulk aerosolmodel (BAM) configuration (Rasch et al., 2001; Lamarqueet al., 2012). The moist turbulence scheme in CAM5 replacesthe dry turbulence scheme in previous versions and explicitlysimulates cloud–radiation–turbulence interactions, allowingfor a more realistic simulation of aerosol semi-direct effectsin stratus clouds (Bretherton and Park, 2009). Also includedin CAM5 are an improved shallow convection scheme anda revised cloud macrophysics scheme (Neale et al., 2010).The atmospheric chemistry component is now fully interac-tive and embedded within CAM5 and handles emissions ofaerosols and trace gases and deposition of aerosols to snow,ice and vegetation. Our simulations did not use the ModalAerosol Model (MAM) to simulate cloud indirect effects(Liu et al., 2012), with efforts still ongoing to improve therepresentation of these processes within CAM. Evaluatingindirect effects on the climate response documented here isan important next step.

To estimate landscape fire emissions, we used gaseousand particulate fire emissions from the Global Fire Emis-sions Database, version 3 (GFEDv3) (van der Werf et al.,2010). Calculation of burned area in GFEDv3 is describedby Giglio et al.(2010). Fuel loads and combustion complete-ness factors are estimated using a biogeochemical model andare combined with satellite-derived burned area estimates toderive total carbon emissions. Aerosol emissions are thenestimated from total emissions using emissions factors fordifferent biomes, drawing upon published emission factorsfrom Andreae and Merlet(2001) that are updated annually.Akagi et al.(2011) have published an update to emission fac-tors fromAndreae and Merlet(2001) that were not availableduring the construction of GFEDv3, but likely will be in-corporated in a future version of the GFED model. We usedthe Multi-angle Imaging SpectroRadiometer (MISR) Level3 daily AOD product (MIL3MAE) and the Moderate Reso-lution Imaging Spectro-radiometer (MODIS) Level 3, Col-lection 5 monthly AOD product (MOD08 M3), to assist inscaling the GFED aerosol emissions.

We used ground-based Aerosol Robotic Network(AERONET) optical depth data (Holben et al., 1998) from21 individual stations to evaluate our model simulationswith adjusted emissions. We assessed the strength andspatial location of the Hadley circulation using horizontaland vertical wind velocities obtained from the European

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013

5230 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

Centre for Medium-Range Weather Forecasting (ECMWF)interim Reanalysis product (ERA-interim) (Dee et al., 2011).Monthly ERA-interim data were available from 1989–2011at a 0.75◦ × 0.75◦ horizontal resolution with 60 verticallevels.

2.2 Scaling fire emissions to achieve realistic AODs

We forced an initial simulation of the CAM5-SOM con-figuration of CESM with monthly varying fire emissionsfrom GFEDv3 during 1997–2009. Evidence fromWardet al. (2012) suggests that to accurately simulate observedaerosol optical depths, GFEDv3 smoke emissions need tobe approximately doubled. Therefore, our initial simula-tions were forced with 2×GFEDv3 BC and OC emissionsand 1×GFEDv3 SO2 emissions. Initialization datasets wereproduced using linear interpolation to re-grid the originalGFEDv3 data (0.5◦ × 0.5◦ spatial resolution) to the CAM5resolution (1.9◦ × 2.5◦). This study isolated the climate re-sponse to aerosols-only; we thus excluded altering nitro-gen emissions as some molecules of NO2 act as precursorsto ozone formation. Aerosols were injected into the lowestlayer of the model, as evidence suggests that smoke injec-tion above the boundary layer is rare (Val Martin et al., 2010;Tosca et al., 2011).

Comparison of the resulting CAM5-simulated AODs toobservations from both MISR and MODIS revealed a lowbias in biomass burning regions (Fig. S1). Some of the biasmay be explained by a lack of an explicit parametrization ofsecondary aerosol condensation and coagulation processes inCAM5. Specifically, the emission factors we used fromAn-dreae and Merlet(2001) may include measurements madeprior to significant plume aging and condensation. Studieshave shown that secondary aerosols constitute a significantfraction of the total aerosol mass within biomass burningplumes (Lee et al., 2008; Grieshop et al., 2009). For exam-ple, organic carbon aerosol concentrations increased by fac-tors of 1.5 to 6 after 3 to 4 h of aging downwind of a pre-scribed fire in Georgia (Lee et al., 2008). It is also likely thatthe GFEDv3 inventory underestimated emissions contribu-tions from small fires by as much as 35 % (Randerson et al.,2012). Furthermore, there is some evidence that liquid cloudfraction and wet deposition rates are too high in CAM5 andthat this contributes to increased wet aerosol deposition andthus low optical depth biases (Wang et al., 2013).

In the three major tropical burning regions of South Amer-ica (SAM), southern Africa (SAF) and equatorial Asia (EAS)(Fig. 1), AODs were substantially lower than observationsfrom MISR and MODIS (Fig. 2). For these regions, andalso for boreal North America, we computed the scalingfactor required to bring the AODs into agreement with thesatellite time series. Our scaling factors apply only to di-rect aerosols emissions as we did not explicitly includeany parametrization of secondary organic aerosol formationwithin fire plumes. We chose regions where fire aerosols

a)

b)

c)

d)

Fig. 1. Regional maps of AERONET stations (black dots) andMISR/MODIS scaling areas (blue boxes) for(a) South Amer-ica (SAM), (b) southern Africa (SAF),(c) equatorial Asia (EAS)and (d) boreal North America (BNA). There were no suitableAERONET stations in BNA.

were the dominant contributor to the optical depth signalwithin CAM5, thereby increasing the likelihood of a mono-tonic relation between emissions and optical depth (Fig. S2).We chose SAM (25◦ S–0; 65–40◦ W), SAF (15–5◦ S; 10–30◦ E), EAS (10◦ S–7◦ N; 90◦ E-150◦ E) and boreal NorthAmerica (BNA; 50–70◦ N; 170–90◦ W) as our initial scal-ing regions. We then derived four regionally-specific meanscale factors by computing the ordinary least squares regres-sion between the simulated AOD (independent variable) andthe observed (dependent variable) for those months in thetime series that cumulatively contributed to 80 % of regional

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5231

a

b

c

MISR and MODIS AOD

CA

M5

AO

D

d

Fig. 2. Linear relations between CAM5-simulated aerosol opticaldepths (y-axis) and MISR-MODIS optical depths (x-axis) for unad-justed case (blue dots/line) and adjusted case (black dots/line) emis-sions cases. Regression slopes for the original emissions (Borig.)and adjusted emissions (Badj.) model simulations are shown ineach panel. The three regions shown are(a) South America (SAM;25◦ S:0, 40:65◦ W), (b) southern Africa (SAF; 15:5◦ S,10:30◦ E)(c) Equatorial Asia (EAS; 10◦ S:7◦ N, 90:150◦ E), and (d) bo-real North America (BNA; 50◦ N:70◦ N, 170:90◦ W). Only thosemonths that cumulatively contributed 80 % of regional emissionsfrom 1997–2009 were included in the analysis. Correlation coeffi-cients (r2) were 0.65 (unadj.) and 0.62 (adj.) for SAM; 0.72 (un-adj.), 0.55 (adj.) for SAF; 0.69 (unadj.) and 0.71 (adj.) for EAS; and0.78 (unadj.) and 0.83 (adj.) for BNA.

fire emissions (Table 1). Each region’s mean scaling factorwas the average of scalars derived separately for MISR andMODIS observations. In other regions, where contributionsfrom other aerosol sources were proportionally larger, it wasnot possible to use this optimization approach. In these re-gions we assigned scale factors based on ecosystem simi-larity and proximity. The scalars for SAM, SAF, EAS andBNA were 2.40, 2.10, 1.67 and 1.45, respectively, and wereapplied to biogeographically similar regions, as shown in Ta-ble 2. In a second simulation we increased emissions by thesescalars, preserving the same spatial and temporal distribu-tions. Global smoke (the sum of BC and OC) emissions fromlandscape fires increased from 40.6 Tg yr−1 to 79.9 Tg yr−1

as a result of the adjustment process. Total SO2 emissionswere adjusted upward from 2.4 to 4.7 Tg yr−1. These ad-justments were broadly similar to estimates fromJohnstonet al. (2012) who applied similar scaling techniques usingthe global GEOS-Chem model to study aerosol effects on hu-man health. Time series biases, root mean squared errors andlinear correlations (slopes) for each region showed consid-erable improvement between the original and adjusted cases(Fig. S3).

The second simulation, using adjusted emissions, pro-duced linear fits between modeled and satellite-observedAODs that had slopes closer to 1.0 (ranging from 0.72 to 1.06for SAM, SAF, EAS and BNA; Fig. 2). We evaluated ouradjustments using AOD data from 21 individual AERONETstations across the tropics (Fig. 1). This confirmed our initialassumption that the relation between AODs and emissionswas mostly linear. We compared CAM5 simulated opticaldepth to observations for only those months when greaterthan 30 % of the optical depth from CAM5 was derivedfrom fire. Even after considering the large spatial-scale mis-matches between the model and the observations, our anal-ysis revealed significant improvement in the linear relationbetween modeled and observed optical depths for individ-ual stations in SAM, SAF and EAS (Fig. 3). Despite generalimprovement between the original and adjusted cases, low-biases still persisted in eastern Africa and parts of equato-rial Asia. This suggests the climate impacts we describe inthe following sections are likely to be conservative. Table 3summarizes the AOD improvements for the simulations weobtained after optimization.

2.3 Effects of fire aerosols on climate using CESM

We used the same configuration of CAM5-SOM (describedin Sect. 2.1) to investigate the simulated climate responseto fire aerosol forcing. We conducted two simulations: onewith no prescribed surface fire aerosol emissions (NOFIRE),but aerosols emissions from all other sources, and onewith surface fire aerosol emissions (FIRE) in addition toall other aerosol sources. Emissions for most species werecompiled and adapted from various sources into a compre-hensive dataset described byLamarque et al.(2010). More

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013

5232 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

Table 1.Summary of scaling factors for selected biomass burning regions.

Region

South America1 southern Africa2 equatorial Asia3 boreal North America4

Original sum of BC and OC emissions5 (Tg yr−1) 3.5 4.8 3.3 1.6Number of months contributing to 80 % of emissions 29 31 18 11(out of 156)MODIS scalar 3.03 2.56 1.75 1.87MODIS correlation (r2) 0.71 0.78 0.67 0.88MISR scalar 1.77 1.63 1.59 1.02MISR correlation (r2) 0.71 0.73 0.71 0.84AVERAGE (MISR and MODIS) scalar 2.40 2.10 1.67 1.45Adjusted sum of BC and OC emissions (Tg yr−1) 8.5 10.0 5.6 2.3

1 South America (SAM), region bounded by 25◦ S–0, 65–40◦ W. 2 Southern Africa (SAF), region bounded by 15–5◦ S, 10–30◦ E.3 Equatorial Asia (EAS), region bounded by 10◦ S–7◦ N, 90–150◦ E.4 Boreal North America (BNA), region bounded by 50–70◦ N, 170–90◦ W.5 The original BC and OC emissions were 2×GFEDv3 (van der Werf et al., 2010).

Table 2.Global BC and OC scalars and emissions from satellite-based optimization.

GFED Region∗ Aerosol emissions scalar Sum of OC and BC Adjusted sum of OCemissions from 2xGFEDv3 and BC emissions

(Tg yr−1) (Tg yr−1)

SHSA 2.40 5.3 12.8NHSA same as SHSA 0.4 1.0CEAM same as SHSA 0.4 1.1SHAF 2.10 10.3 21.6NHAF same as SHAF 8.3 17.4EURO same as SHAF 0.09 0.21AUST same as SHAF 2.4 5.0EQAS 1.67 3.6 6.1SEAS same as EQAS 2.1 3.5CEAS same as EQAS 0.7 1.3MIDE same as EQAS 0.03 0.05BONA 1.45 2.1 3.0BOAS same as BONA 4.5 6.5TENA same as BONA 0.3 0.4global total 1.97 40.6 79.9

∗ GFED regions defined as invan der Werf et al.(2010)SHSA = Southern Hemisphere South America, NHSA = Northern Hemisphere South America,CEAM = central America, SHAF = Southern Hemisphere Africa, NHAF = Northern HemisphereAfrica, EURO = Europe, AUST = Australia, EQAS = equatorial Asia, SEAS = southeast AsiaCEAS = central Asia, MIDE = Middle East, BONA = boreal North America, BOAS = borealAsia, TENA = temperature North America.

specifically, surface emissions of trace gases and aerosolsfrom industrial and natural non-fire sources were based onMOZART-4 emissions described inEmmons et al.(2010).For most species, anthropogenic emissions were from thePOET inventory (Granier et al., 2005), except in Asia whereemissions from the REAS inventory were substituted (Oharaet al., 2007). Fire emissions of BC, OC and SO2 were ob-tained following the approach described in Sect. 2.2 (above).Fire emissions of other minor aerosols and trace gases wereprescribed directly from GFEDv3. The standard configura-tion of the Community Land Model (CLM) automatically

quantifies the radiative forcing associated with black carbondeposition on snow, which proves consequential to the highlatitude climate response.

Each simulation began after a 15 yr spin-up period andlasted for 52 yr. For the FIRE case we forced the model withfour cycles of the adjusted 1997–2009 emissions describedabove. As a result, the FIRE simulation included observedyear-to-year variability in emissions during each cycle. TheNOFIRE simulation was identical to the FIRE simulation butdid not include fire emissions of OC, BC or SO2. The use of52 yr simulations allowed us to quantify fire-induced climate

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5233

Table 3.Comparison of optical depths from simulations with original and adjusted GFEDv3 emissions.

Region Observed Modeled Percent (%)

(original emissions) (adjusted emissions) change

MISR 0.141 0.121 0.152 26South America (SAM) MODIS 0.140 0.124 0.158 27

AERONET∗ 0.301 0.112 0.259 129MISR 0.258 0.189 0.289 53

southern Africa (SAF) MODIS 0.278 0.186 0.287 54AERONET 0.253 0.124 0.207 71MISR 0.160 0.089 0.090 1

equatorial Asia (EAS) MODIS 0.155 0.093 0.095 2AERONET 0.190 0.109 0.151 47MISR 0.124 0.051 0.055 8

boreal North America (BNA) MODIS 0.136 0.058 0.062 7AERONET – – – –

∗ AERONET optical depths are only those where greater than 30 % of the AOD simulated by CAM5 is from fire.Regions are the same as those in Table 1.

responses in a statiscally robust way, given the internal cli-mate variability within each simulation and also the large in-terannual variability of fire emissions that occurred in manyregions.

3 Results

3.1 Spatial and meridional climate response to fireaerosol emissions

The presence of fire aerosols in the FIRE simula-tion produced a global, area-weighted AOD increase of1.5× 10−2

± 0.2× 10−2 (10 %) (Table 4), and large regionalincreases over the middle of central South America, Africaand equatorial Asia (Fig. 4). Remote swaths of open oceanalso exhibited significant AOD increases (between 0.001 and0.01), suggesting that the lifetimes of some fire aerosolswere long enough to allow for long-range transport. In mostcases, the maximum AOD increases occurred over regionsof consistently high fire emissions. For example, over south-ern Africa (15–5◦ S; 10–30◦ E) and South America (25◦ S–0; 65–50◦ E), fires increased annual mean AOD by an area-averaged 0.19± 0.03 (199 %) and 0.08± 0.02 (91 %), re-spectively. Zonally-averaged global AOD increases were at amaximum of 0.06 between 10◦ S and 10◦ N, corresponding toconsistently high fire emissions over Africa and South Amer-ica, with another relative maximum between 50◦ N and 60◦ Nover North American and Eurasian boreal forests (Fig. 5).Optical depth exhibited a clear seasonal cycle and reached azonally-averaged maximum during DJF around 5◦ N (0.11)and during JJA at 5◦ S (0.10).

The total, top-of-atmosphere, direct radiative forcing fromfire aerosols was +0.18± 0.10 W m−2 (Fig. 6a; Table 4). Re-gions of highest positive radiative forcing were generally in

the tropical oceans, corresponding to high AODs, though di-rectly over fire source regions (e.g., central Amazonia, bo-real North America), radiative forcing was slightly negative.In response to the aerosol forcing, globally averaged all-skynet surface shortwave (Snet) decreased by 1.3± 0.2 W m−2

(1 %; Fig. 6b; Table 4). Like AOD, the largest changes oc-curred near or downwind of the major tropical burning re-gions. Area-averaged decreases over southern Africa (forthe same region defined above) and South America (forthe same region defined above) were−19.1 ± 3.2 W m−2

(8 %) and−9.1 ± 1.8 W m−2 (4 %), respectively, with neg-ative anomalies up to−30 W m−2 over some regions withinsouthern Africa. The zonally averaged pattern of Snet anoma-lies closely followed AOD, with the maximum reduction(−5 W m−2) occurring just south of the Equator (Fig. 5).

The combination of increased AOD and reduced surfaceshortwave radiation reduced surface temperature in most ar-eas (0.13± 0.01◦C, Table 4, Fig. 6c). Outside of the in-tertropical convergence zone (ITCZ) in the eastern Pacificand the high-latitude storm tracks, the largest reductions intemperature occurred over the continents. In southern Africa(same region as above), average temperature decreased by0.46± 0.07◦C, and over the southern Amazon (same regionas above) by 0.37± 0.07◦C. Global temperature anoma-lies were at a zonally-averaged minimum at the Equator andnorthward (−0.2◦C) but large reductions also occurred nearthe South Pole. Temperature decreases near the Equator and60◦ N corresponded to a relatively small zonal AOD maxi-mum, suggesting that direct forcing from aerosols at higherlatitudes had a proportionately greater impact. However, thelack of a significant spatial correlation between temperaturechanges andSnet anomalies suggests that direct effects fromsmoke on the local atmosphere and surface radiation budgetwere not responsible for all of the meridional and global tem-perature response.

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013

5234 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

Table 4.Summary of the simulated global climate response to fire aerosols.

Earth System variable NOFIRE (control) FIRE–NOFIRE (C.I.a) % change

Global

Aerosol optical depth 0.15 +0.02 (0.002) +10Top-of-atmosphere radiative forcing (W m−2) −0.47 +0.18 (0.10)Net surface shortwave radiation (W m−2) 155.3 −1.3 (0.2) −1Surface air temperature (◦C) 14.8 −0.13 (0.01)Precipitation (mm d−1) 2.88 −0.03 (0.003) −1Mean maximum annual NHψ (×1010kg s−1)b 8.8 −0.1 (0.1) −1Mean maximum DJF NHψ (×1010kg s−1) 23 −0.3 (0.2) −1Width of NH Hadley Cell (1φ◦) 31.3 +0.4 (0.4) +1

South America (SAM)c

Aerosol optical depth 0.09 +0.08 (0.02) +91Net surface shortwave radiation (W m−2) 215.7 −9.1 (1.8) −4Surface air temperature (◦C) 26.7 −0.37 (0.07)Precipitation (mm d−1) 3.62 −0.08 (0.05) −2

southern Africa (SAF)

Aerosol optical depth 0.10 +0.19 (0.03) +199Net surface shortwave radiation (W m−2) 243.2 −19.1 (3.2) −8Surface air temperature (◦C) 24.0 −0.46 (0.07)Precipitation (mm d−1) 3.32 −0.24 (0.05) −7

a C.I. = 95 % confidence interval (standard error× 1.96).b Defined as the change in the maximum Northern Hemisphereψ (horizontally and vertically varying) between the two simulations.c South America and southern Africa regions as defined in Table 1.

On average, global precipitation decreased 2.9× 10−2±

0.3× 10−2 mm d−1 (1 %) (Table 4), but anomalies showeda complex spatial pattern of large precipitation decreasesat the Equator, slightly smaller decreases in the North-ern Hemisphere storm track and increases between 5 and10◦ N (and between 5 and 10◦ S). Over the main burning re-gions of Africa and South America, precipitation decreased2.4× 10−1

± 0.5× 10−1 mm d−1 (7 %) and 0.8× 10−1±

0.5× 10−1 mm d−1 (2 %), respectively. Some of this precipi-tation decrease appeared to have been caused by local aerosoleffects on surface convergence, upward vertical wind speeds(ω) and atmospheric warming and its effect on the lapse rate.For example, the temperature difference over Africa (sameregion as above) between 700 mb and the surface decreasedby 0.43± 0.10◦C, reflecting increased atmospheric stabil-ity and occurring simultaneously with a decrease in upwardwind velocity of 9.1× 10−4

± 12.7× 10−4 Pa s−1 at 500 mb(Fig. S4). It is likely, however, that other mechanisms areneeded to explain the macroscale change in global precip-itation, including changes in the remote Pacific shown inFig. 4d.

3.2 Fire aerosol effects on the Hadley circulation

We used meridional wind velocities and surface pressure tocompute the annual mean mass streamfunction (described

by Oort and Yienger, 1996) for ERA-interim data and ourCAM5 simulations (Fig. 7a, b). Two Hadley cells, between30◦ S and 30◦ N, were visible in both the ERA-interimdata as well as the CAM5 simulations. The model ade-quately matched the placement and strength of the twocells when compared to the reanalysis. The simulated andobserved streamfunctions (ψ) placed the dividing line be-tween the southern and northern Hadley cells just north ofthe Equator, corresponding to the latitude of mean ascentand near-permanent residence of the ITCZ at 5◦ N. ERA-interim data indicated a slightly stronger southern Hadleycell with maximumψ values exceeding−11× 1010 kg s−1,compared to−8.5× 1010 kg s−1 for CAM5. However, max-imum ψ values for the northern cell were similar betweenmodel and data: 8.1× 1010 kg s−1 vs. 8.8× 1010 kg s−1, re-spectively. Vertical velocity (ω) fields from ERA-interimdata and CAM5 simulations showed the region of maxi-mum ascent (negativeω values) between 10◦ S and 10◦ N,roughly corresponding to the division between the northernand southern Hadley cells (Fig. 8a, b). Upward velocitiesnear 2× 10−2 m s−1 characterized the ascending branches ofthe Hadley cells.

Presence of fire aerosols at the Equator in the FIRE simu-lation weakened both the northern and southern Hadley cells(Fig. 7c, d). The southern Hadley cell increased by as muchas 3.0× 109 kg s−1 around 5◦ S, representing a net reduction

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5235

a

b

c

Fig. 3. Linear relations between CAM5 simulated aerosol opticaldepths (y-axis) and AERONET optical depths (x-axis) for the un-adjusted case (blue dots/line) and adjusted case (black dots/line),showing better agreement in the adjusted scenario. Regions are thesame as Fig. 2:(a) South America (SAM),(b) southern Africa(SAF) and (c) equatorial Asia (EA). Only those months whereCAM5 AOD from fire emissions was greater than or equal to 30 %were used in our comparisons.

in southward transport of around 10 %, though reductionswere smaller further south in the region of maximum ab-soluteψ . Similarly, ψ values in the northern Hadley celldecreased by−3.8× 109 kg s−1 at 5◦ N, also an approxi-mate 10 % reduction in northward transport. The maximumψ for DJF decreased from 2.30× 1011 to 2.27× 1011 kg s−1

(a reduction of 0.3± 0.2× 1010 kg s−1), though reductionsin excess of 6.7× 109 kg s−1 occurred closer to the Equator.Despite Hadley cell weakening, the width of the tropics in-creased slightly. We calculated the annually averaged north-ward extent of the Hadley cell for each simulation as the lat-itude (φ) at whichψ (at 500 mb) switched from positive tonegative, as described inAllen et al.(2012a). We found that

Fig. 4.Global map of aerosol optical depth anomalies (FIRE minusNOFIRE) from the CAM5 simulations. Averages were calculatedusing all 52 yr from each simulation (and excluding the precedingspin-up periods). This applies to all remaining figures and tables.

1φ between the FIRE and NOFIRE cases was 0.4± 0.4◦

suggesting that the tropics widened.Weakening of the Hadley circulation was likely a result of

the aerosol forcing between 10◦ S and 10◦ N (e.g., Fig. 7d).Elevated fire aerosols in this latitude band both cooled thesurface and warmed the atmosphere. In some places, localaerosol-induced subsidence (more positive values ofω) con-tributed to the reduction inψ values near the Equator. Forexample, during the Northern Hemisphere summer (May–October), high AODs over southern Africa contributed to acolumn heating of greater than 0.9 K d−1 from 1000–700 mband local maximum temperature increase of 0.4◦C at 700 mb,both of which increasedω by 4× 10−2 Pa s−1 near 850 mband limited the amount of equatorial convection (Fig. S5).This caused a local weakening of the poleward transport ofmass in the southern Hadley cell.

Similarly, the global reduction in upward vertical veloc-ities near the Equator (and subsequent weakening ofψ ,Fig. 8c) appeared to be linked with sharp reductions inSST and mid-tropospheric heating in a narrow swath be-tween 5◦ S and 5◦ N. In particular, over much of the Pa-cific the largestω increases were co-located with reductionsin SSTs, suggesting that the fire-induced temperature de-creases had the largest effect onω in regions of maximumconvection. Pronounced heating between 1000 and 500 mbsuggested that the long-range transport of aerosols over thePacific contributed to the suppression of convection. Sharpdecreases in atmospheric heating rates at altitudes above500 mb corroborate a reduction in mid- to upper-level con-densation. Over the tropical Pacific (180–90◦ W), ω anoma-lies exceeded 2.0× 10−5 Pa s−1 in response to SST reduc-tions greater than 0.3◦C and maximum heating rates of 0.1 Kd−1 at 850 mb (Fig. S6).

4 Discussion

Simulated fire aerosols reduced net surface shortwaveradiation, especially over the major burning regions ofSouth America, Africa and equatorial Asia, and increased

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013

5236 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

a) TOA direct radiative forcing (W m-2)

c) Surface air temperature (°C)

d) Precipitation (mm day-1)

b) Net surface SW radiation (W m2)

Fig. 5.Zonally averaged climate anomalies (FIRE – NOFIRE) fromCAM5 simulations:(a) aerosol optical depth,(b) net insolation(W m−2), (c) temperature (◦C), and(d) precipitation (percent (%)change). Thin lines are seasonal averages, thick lines are annual av-erages.

atmospheric warming, especially in the tropics and mid-latitudes. Global surface air temperatures were lower and insome places negative anomalies exceeded−0.5◦C. Thoughchanges in surface radiation were largely confined to highbiomass burning regions, the temperature response was moreglobally distributed. This was likely due to a substantial re-duction in heat transport from the tropics to mid- and high-latitude regions. The surface temperature reductions com-bined with increased tropospheric heating near the Equatorreduced convection in the ascending branches of the two

optic

al d

epth

net s

urfa

ce S

W (

W m

-2)

surf

ace

air

tem

pera

ture

(ºC

)

prec

ipita

tion

(% c

hang

e)

a)

b)

c)

d)

Fig. 6. Global maps of climate anomalies (FIRE minus NOFIREfor (b–d) only) from CAM5 simulations:(a) top-of-atmosphere ra-diative forcing (W m−2 from FIRE run only),(b) net insolation(W m−2), (c) surface air temperature (◦C), and(d) precipitation(mm d−1). Dotted stippling of statistical significance (95 %). Sig-nificance was determined by computing thet test statistic at eachgrid cell for α = 0.05. Surface air temperature was the mean mid-layer air temperature in the lowest atmospheric level of the model.

Hadley cells. These results are consistent with conclusionsfrom Tosca et al.(2010) which showed a link between fireemissions and precipitation reductions in equatorial Asia. Insum, the presence of fire aerosols in the troposphere caused asmall general weakening of the northern and southern Hadleycells in simulations with CAM5.

The mechanisms for Hadley cell weakening are alsolargely consistent with results fromQuan et al.(2005) thatlink SSTs to the strength of the Hadley circulation. Theysuggest that from 1950 to present, increased surface tem-peratures have contributed to a gradual strengthening ofthe Hadley circulation. They also note that the strength ofthe Hadley circulation is positively correlated with El Nino(warm SST) events in the eastern Pacific (and negatively cor-related with La Nina (cold SST) events).Mitas and Clement(2005) andLu et al.(2007) also present evidence that surfacewarming is positively correlated with Hadley cell strength.

The latter study found a 50.4× 108 kg s−1 increase in themaximum DJF Northern Hemisphere mass streamfunctionduring 1979–2003, a period when surface temperatures in-creased by 0.6◦C (Hansen et al., 2010). Given a mean valueof 8.8× 1010 kg s−1, this corresponds to a cumulative in-crease of 5.7 %. Although decadal changes in fire emissions

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5237

optic

al d

epth

a) ERA-Interim Obs.

b) CAM5 FIRE

c) CAM5 FIRE – NOFIRE

d)

Fig. 7.Zonal-mean (annual) mass streamfunction (ψ) derived from(a) ECMWF ERA-interim observations,(b) CAM5 simulations in-cluding fire aerosols, and(c) the difference between the FIRE andNOFIRE simulations. Units are in 109 kg s−1 for all plots. Con-tour intervals vary. Shaded regions indicate northward transport, un-shaded regions are southward transport.(d) is the zonally-averagedAOD.

are not well understood, it is likely that deforestation andsavanna woodland fires have increased significantly since1950. For illustrative purposes, if we assume fires increasedby approximately 50 % over this time, then, using a pre-liminary analysis of output from our simulations, fires mayhave offset Hadley strengthening during this interval by4× 108 kg s−1. Thus, in the absence of possible changes inthe fire regime, the strengthening of the Hadley circulationcould have been approximately 8 % greater.

Analysis of reanalysis observations suggests that the widthof the Northern Hemisphere Hadley circulation has increasedin recent decades, by 0.3◦/decade during 1979–1999 (Allenet al., 2012a). Though we simulate a decrease in Hadley cellstrength, we also show a significant widening of the annualnorthern Hadley cell (1φ = 0.4± 0.4◦), in the same directionas the observations. This is consistent with results fromAllenet al.(2012b) who showed that recent observations of Hadleycell expansion can be partly attributed to mid-latitude tropo-spheric heating from black carbon aerosols. Using variousmeasures for determining tropical width, their simulations at-tribute an increase of 0.3–1.0◦ decade−1 for 1979–2009 frommid-latitude BC warming of the lower troposphere. Surfaceair warming from greenhouse gas forcing is known to par-tially explain recent increases in Hadley cell strength, but astronger Hadley circulation usually results in an equatorwardcontraction (Lu et al., 2008). However, black carbon heat-ing increases atmospheric stability, which pushes the baro-clinic zone poleward, resulting in an expansion of the Hadley

a) ERA-Interim obs.

b) CAM5 FIRE

c) CAM5 FIRE – NOFIRE

vert

ical

vel

ocityd)

Fig. 8. Zonal-mean (annual) vertical velocities (ω) derived from(a) ECMWF ERA-interim observations,(b) CAM5 simulations in-cluding fire aerosols, and(c) the difference between the FIRE andNOFIRE simulations. Units are in 10−4 Pa s−1 for all plots. Con-tour intervals vary. Negative values (shaded regions) indicate up-ward velocities, positive values (un-shaded regions) are downwardvelocities.(d) is the 500 mb vertical velocity anomalies (as inc).

cell. Following the same fire scenario as in the previous para-graph, our CESM simulations suggest fires may have con-tributed to approximately 10 % of the observed trend.

Given that we scaled fire emissions to match simulatedAODs to observations in burning regions, it is likely thatour simulations adequately but conservatively captured themagnitude of the direct forcing from fire aerosols. For exam-ple, we estimated that fires increased AOD by approximately0.02, which is in line with estimates of 0.02–0.03 fromMa-howald et al.(2011) and 0.03 fromBauer and Menon(2012).This represents a 10 % increase over the global backgroundaerosol load. We also acknowledge that scaling surface emis-sions so that simulated AODs match observations is not aseamless fix to the underestimation of AOD within CAM5,and that other factors, such as secondary aerosol formationand wet deposition processes (Xian et al., 2009), may con-tribute to discrepancies between simulations and observa-tions.

Our results demonstrate a plausible link between smokeaerosols and changes in global circulation but do not ad-dress whether simulated circulation changes have any impacton fire distribution or occurrence. Elevated AODs generallyreduced surface temperatures, especially those in the tropi-cal Pacific where our simulations showed a La Nina-like re-sponse to the smoke forcing. The combination of decreasedtemperatures, atmospheric heating and aerosol-cloud indi-rect effects reduced convection at the Equator and weakenedthe Hadley circulation. Over some locales, like the tropical

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013

5238 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

forests of Africa and South America, simulated reductionsin precipitation (between 5 and 15 %) lowered soil moisturecontent in the top several layers, which increased droughtstress. This would make it easier for land managers to usefire as a tool in clearing land for pastures, croplands or plan-tations. Combined with the modeled relationship betweenglobal warming and tropical drying (Neelin et al., 2006), theincreased drought stress may enhance positive feedbacks be-tween fire and climate. However, some of these ecosystemimpacts are likely offset or modified by the strengthening ofthe Hadley cells in response to global warming.

Owing to the coarse resolution of CAM5 and the compli-cated relationship between cloud microphysics and aerosols,it is intrinsically difficult to simulate the mesoscale meteo-rological response to smoke. In regions like equatorial Asia,geography and complicated sea-breeze interactions make itdifficult to model convection, and thus difficult to fully real-ize the climate response to smoke-aerosol forcing. We notethe difficulty in accurately representing spatial and temporalpatterns of precipitation and circulation changes. This study,therefore, is a first estimate of the global climate response tofire emissions from CAM5 that accounts for direct and semi-direct aerosol effects.

5 Conclusions

We used a global climate model to simulate the sensitivity ofthe climate to fire aerosols. We first optimized black and or-ganic carbon emissions by matching simulated and observedoptical depths. Validation of modeled AODs with surface-based measurements showed that our emissions yielded morerealistic distributions of aerosols after our scaling approach.Global simulations that included fire emissions produced el-evated AODs, especially across the tropics. In response to theaerosol forcing, global temperatures declined with maximumreductions in the tropics. Changes in precipitation patternssuggest that fire-emitted aerosols modify global circulationthrough a combination of decreased surface insolation, atmo-spheric heating, reduced surface temperature and increasedsubsidence globally and in tropical convective regions. Ourresults suggest a link between fire aerosols and the strengthand extent of the Hadley circulation.

Important next steps include assessing the regional im-pact of fire aerosols, inclusion of indirect effects in model-ing studies and determining the relative importance of thedirect and indirect aerosol contributions to the climate re-sponse. Assessing which regions contribute the most to thelarge response in the eastern Pacific could be done by iso-lating emissions from Africa, South America and other highburning regions in individual simulations. Furthermore, theModal Aerosol Model (MAM) has been developed and em-bedded in the latest version of CAM5 and simulates aerosolindirect effects in stratus clouds (Liu et al., 2012). One im-portant direction for future research is to isolate the individ-

ual contributions from the direct and indirect aerosol effects,using MAM embedded within CAM5. A final important nextstep is understanding the combined effects of fire-inducedchanges in solar radiation, precipitation, albedo and deposi-tion on tropical ecosystem function.

Supplementary material related to this article isavailable online at:http://www.atmos-chem-phys.net/13/5227/2013/acp-13-5227-2013-supplement.pdf.

Acknowledgements.We are grateful for support from NSF(AGS-1048890) and NASA (NNX11AF96G). M.G.T. receivedsupport from a NASA Earth and Space Science Fellowship(NNX08AU90H). C.S.Z. acknowledges NSF (ARC-0714088) andNASA (NNX07AR23G) support.

Edited by: Y. Balkanski

References

Ackerman, A. S., Toon, O. B., Stevens, D. E., Heyms-field, A. J., Ramanathan, V., and Welton, E. J.: Reduc-tion of tropical cloudiness by soot, Science, 288, 1042–1047,doi:10.1126/science.288.5468.1042, 2000.

Akagi, S. K., Yokelson, R. J., Wiedinmyer, C., Alvarado, M. J.,Reid, J. S., Karl, T., Crounse, J. D., and Wennberg, P. O.: Emis-sion factors for open and domestic biomass burning for usein atmospheric models, Atmos. Chem. Phys., 11, 4039–4072,doi:10.5194/acp-11-4039-2011, 2011.

Albrecht, B. A.: Aerosols, cloud microphysics, and fractionalcloudiness, Science, 245, 1227–1230, 1989.

Allen, R. J., Sherwood, S. C., Norris, J. R., and Zender, C. S.: Theequilibrium response to idealized thermal forcings in a compre-hensive GCM: implications for recent tropical expansion, At-mos. Chem. Phys., 12, 4795–4816,doi:10.5194/acp-12-4795-2012, 2012a.

Allen, R. J., Sherwood, S. C., Norris, J. R., and Zender, C. S.: Re-cent Northern Hemisphere tropical expansion primarily drivenby heterogeneous warming agents, Nature, 485, 350–354, 2012b.

Andreae, M. O. and Merlet, P.: Emission of trace gases and aerosolsfrom biomass burning, Glob. Biogeochem. Cy., 15, 955–966,2001.

Andreae, M. O. and Rosenfeld, D.: Aerosol cloud precipitationinteractions. Part 1. The nature and sources of cloud-activeaerosols, Earth Sci. Rev., 89, 13–41, 2008.

Andreae, M. O., Rosenfeld, D., Artaxo, P., Costa, A. A., Frank,G. P., Longo, K. M., and Silva-Dias, M. A. F.: Smok-ing rain clouds over the Amazon, Science, 303, 1337–1342,doi:10.1126/science.1092779, 2004.

Bauer, S. E. and Menon, S.: Aerosol direct, indirect, semidi-rect, and surface albedo effects from sector contributionsbased on the IPCC AR5 emissions for preindustrial andpresent-day conditions, J. Geophys. Res., 117, D01206,doi:10.1029/2011JD016816, 2012.

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5239

Bowman, D. M. J. S., Balch, J. K., Ataxo, P., Bond, W. J., Carlson,J. M., Cochrane, M. A., D’Antonio, C. M., DeFries, R. S., Doyle,J. C., Harrison, S. P., Johnston, F. H., Keeley, J. E., Krawchuk,M. A., Kull, C. A., Marston, J. B., Moritz, M. A., Prentice, I. C.,Roos, C. I., Scott, A. C., Swetnam, T. W., van der Werf, G. R.,and Pyne, S. J.: Fire in the Earth System, Science, 324, 481–484,doi:10.1126/science.1163886, 2009.

Bretherton, C. S. and Park, S.: A new moist turbulence parametriza-tion in the Community Atmosphere Model, J. Clim., 22, 3422–3448, 2009.

Chen, Y., Randerson, J. T., Morton, D. C., DeFries, R. S., Col-latz, G. J., Kasibhatla, P. S., Giglio, L., Jin, Y., and Marlier,M. E.: Forecasting fire season severity in South America us-ing sea surface temperature anomalies, Science, 334, 787–791,doi:10.1126/science.1209472, 2011.

Chylek, P. and Wong, J.: Effect of absorbing aerosols on global ra-diation budget, Geophys. Res. Lett., 8, 929–931, 1995.

Collins, W. D., Rasch, P. J., Boville, A., Hack, J. J., MaCaa, J. R.,Williamson, D. L., Kiehl, J. T., Briegleb, B. P., Bitz, C., Lin, S. J.,Zhang, M., and Dai, Y.: Description of the NCAR CommunityAtmosphere Model (CAM 3.0), Tech. rep., National Center forAtmospheric Research, 2004.

Crutzen, P. J., Heidt, L. E., Krasnec, J. P., Pollock, W. H., and Seiler,W.: Biomass burning as a source of atmospheric gases CO, H2N2O, NO, CH3Cl and COS, Nature, 282, 253–256, 1979.

Dee, D. P., Uppala, S. M., Simmons, A. J., Berrisford, P., Poli,P., Kobayashi, S., Andrae, U., Balmaseda, M. A., Balsamo, G.,Bauer, P., Bechtold, P., Beljaars, A. C. M., van de Berg, L., Bid-lot, J., Bormann, N., Delsol, C., Dragani, R., FUentes, M., Geer,A. J., Haimberger, L., Healy, S. B., Hersbach, H., Holm, E. V.,Isaksen, L., Kallberg, P., Kohler, M., Matricardi, M., McNally,A. P., Monge-Sanz, B. M., Morcrette, J.-J., Park, B.-K., Peubey,C., de Rosnay, P., Tavolato, C., Thepaut, J.-N., and Vitart, F.: TheERA-Interim reanalysis: configuration and performance of thedata assimilation system, Q. J. Roy. Meteorol. Soc., 137, 553–597, 2011.

Emmons, L. K., Walters, S., Hess, P. G., Lamarque, J.-F., Pfister,G. G., Fillmore, D., Granier, C., Guenther, A., Kinnison, D.,Laepple, T., Orlando, J., Tie, X., Tyndall, G., Wiedinmyer, C.,Baughcum, S. L., and Kloster, S.: Description and evaluation ofthe Model for Ozone and Related chemical Tracers, version 4(MOZART-4), Geosci. Model Dev., 3, 43–67,doi:10.5194/gmd-3-43-2010, 2010.

Feingold, G., Remer, L. A., Ramaprasad, J., and Kaufman, Y. J.:Analysis of smoke impact on clouds in Brazilian biomass burn-ing regions: An extension of Twomey’s approach, J. Geophys.Res., 106, 22907–22922, 2001.

Fernandes, K., Baethgen, W., Bernardes, S., DeFries, R., DeWitt,D. G., Goddard, L., Lavado, W., Lee, D. E., Padoch, C., Pinedo-Vasquez, M., and Uriarte, M.: North tropical Atlantic influenceon western Amazon fire season variability, Geophys. Res. Lett.,38, L12701,doi:10.1029/2011GL047392, 2011.

Field, R. D., van der Werf, G. R., and Shen, S. S. P.: Human ampli-fication of drought-induced biomass burning in Indonesia since1960, Nature Geo., 2, 185–188,doi:10.1038/NGEO443, 2009.

Giglio, L., Randerson, J. T., van der Werf, G. R., Kasibhatla, P.S., Collatz, G. J., Morton, D. C., and DeFries, R. S.: Assess-ing variability and long-term trends in burned area by mergingmultiple satellite fire products, Biogeosciences, 7, 1171–1186,

doi:10.5194/bg-7-1171-2010, 2010.Granier, C., Lamarque, J. F., Mieville, A., Muller, J. F., Olivier,

J., Orlando, J., Peters, J., Petron, G., Tyndall, G., and Wallens,S.: POET, a database of surface emissions of ozone precursors,available at:http://www.aero.jussieu.fr/projet/ACCENT/POET.php, 2005.

Grieshop, A. P., Logue, J. M., Donahue, N. M., and Robinson, A.L.: Laboratory investigation of photochemical oxidation of or-ganic aerosol from wood fires 1: measurement and simulation oforganic aerosol evolution, Atmos. Chem. Phys., 9, 1263–1277,doi:10.5194/acp-9-1263-2009, 2009.

Gunn, R. and Phillips, B. B.: An experimental investi-gation of the effect of air pollution on the initiationof rain, J. Meteor., 14, 272–280,doi:10.1175/1520-0469(1957)014¡0272:AEIOTE¿2.0.CO;2, 1957.

Hansen, J., Sato, M., and Ruedy, R.: Radiative forcing and climateresponse, J. Geophys. Res., 102, 6831–6864, 1997.

Hansen, J., Ruedy, R., Sato, M., and Lo, K.: Global sur-face temperature change, Rev. Geophys., 48, RG4004,doi:10.1029/2010RG000345, 2010.

Held, I. M. and Soden, B. J.: Robust responses of the hydro-logical cycle to global warming, J. Clim., 19, 5686–5699,doi:10.1175/2010JCLI4045.1, 2006.

Holben, B. N., Eck, T. F., Slutsker, I., Tanre, D., p. Buis, J., Set-zer, A., Vermote, E., Reagan, J. A., Kaufman, Y., Makajima, T.,Lavenu, F., Jankowiak, I., and Smirnov, A.: AERONET – A fed-erated instrument network and data archive for aerosol character-ization, Remote Sens. Environ., 66, 1–16, 1998.

Johnston, F. H., Henderson, S. B., Chen, Y., Randerson, J. T.,Marlier, M., DeFries, R. S., Kinney, P., Bowman, D. M.J. S., and Brauer, M.: Estimated global mortality attributable tosmoke from landscape fires, Environ. Health Per., 120, 695–701,doi:10.1289/ehp.1104422, 2012.

Jones, A., Haywood, J. M., and Boucher, O.: Aerosol forc-ing, climate response and climate sensitivity in the HadleyCentre climate model, J. Geophy. Res., 112, D20211,doi:10.1029/2007JD008688, 2007.

Kaufman, Y. J., Fraser, R. S., and Mahoney, R. L.: Fossil fuel andbiomass burning effect on climate–heating or cooling?, J. Cli-mate, 4, 578–588, 1991.

Koren, I., Kaufman, Y. J., Remer, L. A., and Martins, J. V.: Measure-ment of the effect of Amazon smoke on inhibition of cloud for-mation, Science, 303, 1342–1345,doi:10.1126/science.1089424,2004.

Koren, I., Kaufman, Y. J., Rosenfeld, D., Remer, L. A., and Rudich,Y.: Aerosol invigoration and restructuring of Atlantic clouds,Geophys. Res. Lett., 32, L14828,doi:10.1029/2005GL023187,2005.

Lamarque, J.-F., Bond, T. C., Eyring, V., Granier, C., Heil, A.,Klimont, Z., Lee, D., Liousse, C., Mieville, A., Owen, B.,Schultz, M. G., Shindell, D., Smith, S. J., Stehfest, E., Van Aar-denne, J., Cooper, O. R., Kainuma, M., Mahowald, N., Mc-Connell, J. R., Naik, V., Riahi, K., and van Vuuren, D. P.: His-torical (1850–2000) gridded anthropogenic and biomass burningemissions of reactive gases and aerosols: methodology and ap-plication, Atmos. Chem. Phys., 10, 7017–7039,doi:10.5194/acp-10-7017-2010, 2010.

Lamarque, J.-F., Emmons, L. K., Hess, P. G., Kinnison, D. E.,Tilmes, S., Vitt, F., Heald, C. L., Holland, E. A., Lauritzen,

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013

5240 M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires

P. H., Neu, J., Orlando, J. J., Rasch, P. J., and Tyndall, G.K.: CAM-chem: description and evaluation of interactive at-mospheric chemistry in the Community Earth System Model,Geosci. Model Dev., 5, 369–411,doi:10.5194/gmd-5-369-2012,2012.

Langenfelds, R. L., Francey, R. J., Pak, B. C., Steele, L. P., Lloyd,J., Trudinger, C. M., and Allison, C. E.: Interannual growth ratevariations of atmospheric CO2 and itsδ13C, H2, CH4, and CObetween 1992 and 1999 linked to biomass burning , Global Bio-geochem. Cy., 16, 1048,doi:10.1029/2001GB001466, 2002.

Lee, S., Kim, H. K., Yan, B., Cobb, C. E., Hennigan, C., Nichols, S.,Chamber, M., Edgerton, E. S., Jansen, J. J., Hu, Y., Zheng, M.,Weber, R. J., and Russell, A. G.: Diagnosis of aged prescribedburning plumes impacting an urban area, Environ. Sci. Technol.,42, 1438–1444,doi:10.1021/es7023059, 2008.

Liu, H. and Randerson, J. T.: Interannual variability of sur-face energy exchange depends on stand age in a borealforest fire chronosequence, J. Geophys. Res., 113, G01006,doi:10.1029/2007JG000483, 2008.

Liu, X., Easter, R. C., Ghan, S. J., Zaveri, R., Rasch, P., Shi, X.,Lamarque, J.-F., Gettelman, A., Morrison, H., Vitt, F., Conley,A., Park, S., Neale, R., Hannay, C., Ekman, A. M. L., Hess, P.,Mahowald, N., Collins, W., Iacono, M. J., Bretherton, C. S., Flan-ner, M. G., and Mitchell, D.: Toward a minimal representationof aerosols in climate models: description and evaluation in theCommunity Atmosphere Model CAM5, Geosci. Model Dev., 5,709–739,doi:10.5194/gmd-5-709-2012, 2012.

Lu, J., Vecchi, G. A., and Reichler, T.: Expansion of the Hadleycell under global warming, Geophys. Res. Lett, 34, L06805,doi:10.1029/2006GL028443, 2007.

Lu, J., Chen, G., and Frierson, D. M. W.: Response of the zonalmean atmospheric circulation to El Nino versus global warming,J. Climate, 21, 5835–5851,doi:10.1175/2008JCLI2200.1, 2008.

Lyons, E. A., Jin, Y., and Randerson, J. T.: Changes in surfacealbedo after fire in boreal forest ecosystem of interior Alaskaassessed using MODIS satellite observations, J. Geophys. Res.,113, G02012,doi:10.1029/2007JG000606, 2008.

Mahowald, N., Ward, D. S., Kloster, S., Flanner, M. G., Heald,C. L., Heavens, N. G., Hess, P. G., Lamarque, J.-F., and Chuang,P. Y.: Aerosol impacts on climate and biogeochemistry, Annu.Rev. Environ. Resour., 36, 45–74,doi:10.1146/annurev-environ-042009-094507, 2011.

Marlon, J. R., Bartlein, P. J., Carcaillet, C., Gavin, D. G., Har-rison, S. P., Higuera, P. E., Joos, F., Power, M. J., and Pren-tice, I. C.: Climate and human influences on global biomassburning over the past two millenia, Nature Geosci., 1, 697–702,doi:10.1038/ngeo313, 2008.

Mitas, C. M. and Clement, A.: Has the Hadley cell been strength-ening in recent decades?, Geophys. Res. Lett., 32, L03809,doi:10.1029/2004GL021765, 2005.

Morton, D. C., DeFries, R. S., Nagol Jr., J., C. M. S., Kasischke,E. S., Hurtt, G. C., and Dubayah, R.: Mapping canopy damagefrom understory fires in Amazon forests using annual time seriesof Landsat and MODIS data, Remote Sens. Environ., 115, 1706–1720,doi:10.1016/j.rse.2011.03.002, 2011.

Myhre, G., Govaerts, Y., Haywood, J. M., Berntsen, T. K.,and Lattanzio, A.: Radiative effect of surface albedo changefrom biomass burning, Geophys. Res. Lett., 32, L20812,doi:10.1029/2005GL022897, 2005.

Neale, R. B., Chen, C.-C., Gettelman, A., Lauritzen, P. H., Park,S., Williamson, D. L., Conley, A. J., Garcia, R., Kinnison, D.,Lamarque, J.-F., Marsh, D., Mills, M., Smith, A. K., Tilmes, S.,Vitt, F., Morrison, H., Cameron-Smith, P., Collins, W. D., Iacono,M. J., Easter, R. C., Liu, X., and Taylor, M. A.: Description of theNCAR Community Atmosphere Model (CAM 5.0), Tech. rep.,National Center for Atmospheric Research, 2010.

Neelin, J. D., Munnich, M., Su, H., Meyerson, J. E., and Hol-loway, C. E.: Tropical drying trends in global warming mod-els and observations, P. Natl. Acad. Sci., 103, 6110–6115,doi:10.1073/pnas.0601798103, 2006.

Ohara, T., Akimoto, H., Kurokawa, J., Horii, N., Yamaji, K., Yan,X., and Hayasaka, T.: An Asian emission inventory of an-thropogenic emission sources for the period 1980–2020, At-mos. Chem. Phys., 7, 4419–4444,doi:10.5194/acp-7-4419-2007,2007.

Oort, A. H. and Yienger, J. J.: Observed interannual variability inthe Hadley Circulation and its connection to ENSO, J. Climate,9, 2751–2767, 1996.

Page, S. E., Siegert, F., Rieley, J. O., Boehm, H. D., Jaya, A.,and Limin, S.: The amount of carbon released from peat andforest fires in Indonesia during 1997, Nature, 420, 61–65,doi:10.1038/nature01131, 2002.

Penner, J. E., Dickison, R. E., and O’Neill, C. A.: Effects of aerosolfrom biomass burning on the global radiation budget, Science,256, 1432–1434,doi:10.1126/science.256.5062.1432, 1992.

Power, M. J., Marlon, J., Ortiz, N., Bartlein, P. J., Harrison, S. P.,Mayle, F. E., Ballouche, A., Bradshaw, R. H. W., Carcaillet,C., and Cordova, C.: Changes in fire regimes since the LastGlacial Maximum: an assessment based on a global synthe-sis and analysis of charcoal data, Clim. Dyn., 30, 887–907,doi:10.1007/s00382-007-0334-x, 2008.

Quan, X.-W., Diaz, H. F., and Hoerling, M. P.: Change of the tropi-cal Hadley cell since 1950, Kluwer Academic Publishers, 2005.

Ramanathan, V. and Carmichael, G.: Global and regional cli-mate changes due to black carbon, Nature Geosci., 1, 221–227,doi:10.1038/ngeo156, 2008.

Ramanathan, V., Crutzen, P. J., Kiehl, J. T., and Rosenfeld, D.:Aerosols, climate, and the hydrological cycle, Science, 294,2119–2124, 2001.

Randerson, J. T., Chen, Y., van der Werf, G. R., Rogers, B. M.,and Morton, D. C.: Global burned area and biomass burningemissions from small fires, J. Geophys. Res., 117, G04012,doi:10.1029/2012JG002128, 2012.

Rasch, P. J., Collins, W. D., and Eaton, B. E.: Understanding theIndian Ocean Experiment (INDOEX) aerosol distributions withan aerosol assimilation, J. Geophys. Res., 106, 7337–7355, 2001.

Reid, J. S., Koppmann, R., Eck, T. F., and Eleuterio, D. P.: A reviewof biomass burning emissions part II: intensive physical proper-ties of biomass burning particles, Atmos. Chem. Phys., 5, 799–825,doi:10.5194/acp-5-799-2005, 2005.

Reid, J. S., Hyer, E. J., Prins, E. M., Westphal, D. L., Zhang, J.,Wang, J., Christopher, S. A., Curtis, C. A., Schmidt, C. C., Eleu-terio, D. P., Richardson, K. A., and Hoffman, J. P.: Global mon-itoring and forecasting of biomass-burning smoke: Descriptionof and lessons from the Fire Locating and Modeling of BurningEmissions (FLAMBE) program, IEEE J. Appl. Earth. Obs. Rem.Sens., 2, 144–162, 2009.

Atmos. Chem. Phys., 13, 5227–5241, 2013 www.atmos-chem-phys.net/13/5227/2013/

M. G. Tosca et al.: Global climate impacts of smoke aerosols from landscape fires 5241

Rosenfeld, D.: TRMM observed first direct evidence of smoke fromforest fires inhibiting rainfall, Geophys. Res. Lett., 26, 3105,doi:10.1029/1999GL006066, 1999.

Rosenfeld, D.: Aerosols, clouds, and climate, Science, 312, 1323–1324,doi:10.1126/science.1128972, 2006.

Rosenfeld, D., Woodley, W. L., Axisa, D., Freud, E., Hudson, J. G.,and Givati, A.: Aircraft measurements of the impacts of pollutionaerosols on clouds and precipitation over the Sierra Nevada, J.Geophys. Res., 113, D15203,doi:10.1029/2007JD009544, 2008.

Sitch, S., Cox, P. M., Collins, W. J., and Huntingford, C.:Indirect radiative forcing of climate change through ozoneeffects on the land-carbon sink, Nature, 448, 791–794,doi:10.1038/nature06059, 2007.

Spessa, A., McBeth, B., and Prentice, C.: Relationships amongfire frequency, rainfall and vegetation patterns in the wet-drytropics of northern Australia: an analysis based on NOAA-AVHRR data, Global Ecology and Biogeography, 14, 439–454,doi:10.1111/j.1466-822x.2005.00174.x, 2005.

Swetnam, T. W. and Betancourt, J. L.: Mesoscale disturbance andecological response to decadal climatic variability in the Ameri-can Southwest, J. Climate, 11, 3128–3147, 1998.

Tosca, M. G., Randerson, J. T., Zender, C. S., Flanner, M. G., andRasch, P. J.: Do biomass burning aerosols intensify drought inequatorial Asia during El Nino?, Atmos. Chem. Phys., 10, 3515–3528,doi:10.5194/acp-10-3515-2010, 2010.

Tosca, M. G., Randerson, J. T., Zender, C. S., Nelson, D. L., Diner,D. J., and Logan, J. A.: Dynamics of fire plumes and smokeclouds associated with peat and deforestation fires in Indone-sia, J. Geophys. Res., 116, D08207,doi:10.1029/2010JD015148,2011.

Turetsky, M. R., Kane, E. S., Harden, J. W., Ottmar, R. D., and Ma-nies, K. L.: Recent acceleration of biomass burning and carbonlosses in Alaskan forests and peatlands, Nature Geosci., 4, 27–31, 2011.

Val Martin, M., Logan, J. A., Kahn, R. A., Leung, F.-Y., Nelson,D. L., and Diner, D. J.: Smoke injection heights from fires inNorth America: analysis of 5 years of satellite observations, At-mos. Chem. Phys., 10, 1491–1510,doi:10.5194/acp-10-1491-2010, 2010.

van der Werf, G. R., Dempewolf, J., Trigg, S. N., Randerson, J. T.,Giglio, L., Murdiyarso, D., Peters, W., Morton, D. C., Collatz,G. J., Dolman, A. J., and DeFries, R. S.: Climate regulation offire emissions and deforestation in equatorial Asia, P. Natl. Acad.Sci., 105, 20350–20355,doi:10.1073/pnas.0803375105, 2008.

van der Werf, G. R., Randerson, J. T., Giglio, L., Collatz, G. J., Mu,M., Kasibhatla, P. S., Morton, D. C., DeFries, R. S., Jin, Y., andvan Leeuwen, T. T.: Global fire emissions and the contribution ofdeforestation, savanna, forest, agricultural, and peat fires (1997–2009), Atmos. Chem. Phys., 10, 11707–11735,doi:10.5194/acp-10-11707-2010, 2010.

Wang, H., Easter, R. C., Rasch, P. J., Wang, M., Liu, X., Ghan, S.J., Qian, Y., Yoon, J.-H., Ma, P.-L., and Velu, V.: Sensitivity ofremote aerosol distributions to representation of cloud-aerosolinteractions in a global climate model, Geosci. Model Dev. Dis-cuss., 6, 331–378,doi:10.5194/gmdd-6-331-2013, 2013.

Ward, D. S., Kloster, S., Mahowald, N. M., Rogers, B. M., Ran-derson, J. T., and Hess, P. G.: The changing radiative forc-ing of fires: global model estimates for past, present and fu-ture, Atmos. Chem. Phys., 12, 10857–10886,doi:10.5194/acp-12-10857-2012, 2012.

Westerling, A. L., Hidalgo, H. G., Cayan, D. R., and Swet-nam, T. W.: Warming and earlier spring increase west-ern U.S. forest wildfire activity, Science, 313, 940–943,doi:10.1126/science.1128834, 2006.

Xian, P., Reid, J., Turk, J., Hyer, E., and Westphal, D.: Impact ofmodeled versus satellite measured tropical precipitation on re-gional smoke optical thickness in an aerosol transport model,Geophys. Res. Lett., 36, L16805,doi:10.1029/2009GL038823,2009.

Yoshimori, M. and Broccoli, A. J.: On the link between Hadleycirculation changes and radiative feedback processes, Geophys.Res. Lett., 36, L20703,doi:10.1029/2009GL040488, 2009.

Zhang, Y., Fu, R., Yu, H., Qian, Y., Dickinson, R., Dias, M. A. F. S.,da Silva Dias, P. L., and Fernandes, K.: Impact of biomass burn-ing aerosol on the monsoon circulation transition over Amazonia,Geophys. Res. Lett., 36, L10814,doi:10.1029/2009GL037180,2009.

www.atmos-chem-phys.net/13/5227/2013/ Atmos. Chem. Phys., 13, 5227–5241, 2013